- Details
- Parent Category: Programming Assignments' Solutions

# We Helped With This R Programming Assignment: Have A Similar One?

Category | Programming |
---|---|

Subject | R | R Studio |

Difficulty | College |

Status | Solved |

More Info | Statistic Homework Help |

## Short Assignment Requirements

## Assignment Description

Exam12018

*Mark F Owens*

*September 26, 2018*

### Instructions

Please start with this exam template and create a new file in R Markdown called “exam1_your_name” to answer the following questions. Please change the name of the author as well. The questions in Part 1 do not require use of R, but you can type your answers into this R Markdown file if you would like. Part 2 requires you to use R. When you are finished, please knit the file and upload it to Canvas.

### Part 1 (40 Total Points)

**1. **Suppose you want to investigate the factors which
predict a student’s grade on the econometrics exam.

(a) Which factors that your instructor can observe do you think are important? (3 points)

(b) Are there other important factors that your instructor cannot observe? (3 points)

**2. **Suppose you have daily sales data and weather
reports from 2000-2018 for a restaurant. You want to determine if sales are
related to the number of sunny days.

(a) What kind of data do you have? (2 points)

(b) Which is the dependent variable and which is the independent variable in this relationship? Briefly explain.(3 points)

**3. **Specifically,
what does it mean to have a coefficient with a p-value such that p<0.05? (3
points)

**4. **Let *colgpa *denote the grade point average for a college student and let *hsgpa *denote the student’s gpa in high school. A simple model relating college
gpa to high school gpa is

*colgpa *= *β*_{0 }+ *β*_{1}*hsgpa *+ *µ,*

in which *µ *represents the
unobserved error.

(a) What
factors are contained in *µ*? (3 points)

(b) Would
you expect sign of *β*^{ˆ}_{1 }to be positive, negative, or zero? Explain your answer. (3 points)

(c) What
characteristics would the *hsgpa *variable need to have in order for this
simple regression model to uncover the “ceteris paribus” causal effect of *hsgpa *on *colgpa*? Explain. (3 points)

**5. **You are given the following table of output
pertaining to voters in a sample of urban voting districts across the country.
Voters is the number of voters who cast a ballot in 100s, and TV ads is the
number of political advertisements purchased for TV in the market.

Table 1: OLS regression TV advertisements and voters

*Dependent variable:*

| Voters |

TV Ads | 1.25 |

Constant 20.6 |

(a) Use the information in the table to write the regression equation. (3 points)

(b) According to the results, how many voters are generated from each TV advertisement? (3 points)

(c) How much of the variation in voters is explained by this model? Interpret this finding. (3 points)

(d) Suppose one of the voting districts is Erie. There are 30 TV ads in Erie, and 6000 voters. Use the output in the table to calculate the predicted number of voters. (2 points)

(e) What is the residual for Erie? (2 points)

(f) Did Erie have more or fewer voters than would be expected for the number of TV Ads? (2 points)

(g) What other factors would enter into the residual of this regression? (2 points)

**Part 2: Data questions using R (60
Total Points)**

**Step 1: load the data (10 points)**

Please use the data called “micro_grades_2015.csv” to answer questions in Part 2. (10 points)

### Step 2: Summary statistics (20 points)

Use R to calculate the following summary statistics.

(a) What are the mean, median, minimum, and maximum score for *exam1 *for the entire sample? (1 point)

(b) What is the standard deviation of *exam1 *for the entire
sample? (1 point)

(c) What are the mean, median, minimum, and maximum score for *exam2 *for the entire sample? (1 point) (d) What is the standard deviation of *exam2 *for the entire sample? (1 point)

(e) What are the mean, median, minimum, and maximum score for *finalexam *for the entire sample? (1 point)

(f) What is the standard deviation of *finalexam *for the entire
sample? (1 point)

(g) Create a histogram for *exam1 *for the entire sample. Label
the figure. (2 points)

(h) Create a kernel density plot for *exam1 *for the entire
sample. Label the figure. (2 points)

(i) Create a histogram for *exam2 *for the entire sample. Label
the figure. (2 points)

(j) Create a kernel density plot for *exam2 *for the entire
sample. Label the figure. (2 points)

(k) Create a histogram for *finalexam *for the entire sample.
Label the figure. (2 points)

(l) Create a kernel density plot for *finalexam *for the entire
sample. Label the figure. (2 points)

(m) Create
a scatter plot with *exam1 *on the horizontal axis and *finalexam *on
the vertical axis. Add a linear trend. Label the figure. (2 points)

### Step 3 (25 points)

Now use R to create new variables.

(a) Create a new variable called *overallgrade *which takes the
average of *quiz*, *exam1*, *exam2*, and *finalexam *and
include it in your data frame.(3 points)

(b) What are the mean, median, minimum, and maximum of *overallgrade*.
(1 point)

(c) What are the mean, median, minimum, and maximum of *overallgrade *for students who scored 80 or more on *exam1*? (3 points)

(d) Create a new variable called *woverallgrade *which takes a
weighted average of *quiz*, *exam1*, *exam2*, and *finalexam *and
include it in your data frame. The weights are *quiz *10%, *exam1 *25%, *exam2 *25%, and *finalexam *40%. (3 points)

(e) What are the mean, median, minimum, and maximum of *woverallgrade*.
(1 point)

(f) Perform a t-test of whether *overallgrade *is equal to *woverallgrade*.
According to the t-test, are the two measures of overall grade different in a
statistical sense? (3 points)

(g) Perform a t-test of whether *overallgrade *is equal to *exam1*.
According to the t-test, could we just use the score on exam 1 to assign an
overall grade? (3 points)

(h) Create a scatter plot with *exam1 *on the horizontal axis
and *overallgrade *on the vertical axis. Add a linear trend. Label the
figure. (3 points)

(i) Create
a new scatter plot with *exam1 *on the horizontal axis and *overallgrade *on the vertical axis, but do so only for students who scored 80 or higher
on *exam1 *and *exam2*. Add a linear trend. Label the figure. (5
points)

### Part 4: Knit your file as .docx (5 points)

Knit your file as in docx format. (5 points)